Traditionally spanning tree (a spanning tree protocol) has been use to provide both Layer 1 and 2 redundancy, in switch networks, by configuring alternate paths. When a link or switch element fails, a backup link is activated. The spanning tree algorithm defines the active and backup links in the topology. Spanning tree is designed to control the many switches of the topology and has a slow (in the tens of seconds) reconfiguration time.
Spanning tree has a significant disadvantage as to the reconfiguration time. As network systems become more complex and handle additional traffic, the reconfiguration time becomes increasingly problematic.